Put all occurrences of string on new line, then for each line only print specific parameters with their values

Question

I have the following string:

{"attributes":{"name":"James Franco","sectors":{"all":{"name":"all","rank":1},"second":{"name":"Actor","rank":1},"third":{"name":"all","rank":1},"team":"BR",{"name":"all","rank":1},"rank":1,"trending":101,rank":1,"BPR":6.50,"SPR":6.10}}

I want to only extract the 'name', 'BPR' & 'SPR' and their respective values from the one line, and output all three to one line, in a new file.

Can anyone tell me please, how to do this using either gnu sed, cut etc ?

Is there a particular reason to use only "intrinsic" bash commands? — Riccardo Petraglia, May 29 '17 at 21:38
https://stackoverflow.com/questions/1955505/parsing-json-with-unix-tools#1955555 — Hackerman, May 29 '17 at 21:38
The JSON has got quite a few errors in it, so parsing it as it stands is going to be problematic. — Nick, May 29 '17 at 21:39
`"attributes":"name":"James Franco"`... Something appears wrong here... I think it was `"attributes":{"name":"James Franco" .... }` — anishsane, May 30 '17 at 04:51
Apologies - anishsane is correct. I wish to extract the real name, i.e. James Franco, along with the BPR & SPR. — LLB3000, May 31 '17 at 20:19

score 0 · Answer 1 · answered May 29 '17 at 23:49

0

With gnu awk:

$ awk -F: -v RS="," '{gsub(/[}]/,"");gsub("\n","")} \
NR==1{printf "%s:%s%s",$2,$3,RT} $1=="\"BPR\"" || $1=="\"SPR\""{printf "%s%s",$0,RT} \
END{printf "\n"}' file
#Output
"name":"James Franco","BPR":6.50,"SPR":6.10

answered May 29 '17 at 23:49

George Vasiliou

6,130
2
20
27

Thanks George. Can you elaborate on what exactly that command is doing, breaking it down for me please? I have a basic grasp of awk, but this looks quite in depth for me to understand! Thanks. – LLB3000 May 31 '17 at 20:28

Put all occurrences of string on new line, then for each line only print specific parameters with their values

1 Answers1